Sinonasal Side Effects of Chemotherapy and/or Radiation Therapy for Head and Neck Cancer: A Literature Review

Simple Summary Rhinosinusitis and smell alterations are common side effects during and after radiotherapy and chemotherapy for head and neck cancer. The assessment of sinonasal complaints is important to increase patients’ quality of life. The aim of this review is to summarize and analyze our current knowledge of the sinonasal side effects of chemotherapy and/or radiation therapy for head and neck cancer, with a specific focus on mucosal and olfactory disorders. Abstract Radiotherapy and chemotherapy represent important treatment modalities for head and neck cancer. Rhinosinusitis and smell alterations are common side effects in the sinonasal region. This review will summarize and analyze our current knowledge of the sinonasal side effects of chemotherapy and/or radiation therapy for head and neck cancer (HNC), with a specific focus on mucosal and olfactory disorders. A review of the English literature was performed using several databases (PubMed, Embase, Cochrane, Scopus). Fifty-six articles were included in qualitative synthesis: 28 assessed mucosal disorders (rhinitis or rhinosinusitis), 26 evaluated olfactory alterations, and 2 articles addressed both topics. The incidence and severity of olfactory dysfunction and chronic rhinosinusitis were highest at the end of radiotherapy and at three months after treatment and decreased gradually over time. Smell acuity deterioration and chronic rhinosinusitis seemed to be related to radiation dose on olfactory area and nasal cavities, but different degrees of recovery were observed. In conclusion, it is important to establish the severity of chronic rhinosinusitis and olfactory dysfunction in order to find strategies to support patients and improve their quality of life.


Introduction
Head and neck cancer (HNC) is the sixth most common type of cancer worldwide and consists of a heterogeneous group of malignancies that may determine important morbidity to affected patients [1,2]. Smoking and alcohol consumption represent the major risk factors for the development of squamous cell carcinoma (SCC) in the head and neck district. Human papillomavirus (HPV) infection plays a role as an etiologic factor of oropharyngeal SCC, especially in tonsils and the tongue base [3].
HNC treatment includes surgery, radiotherapy (RT), and chemotherapy (CT), which are employed according to the tumor stage and primary site involved [4]. The management The included studies were published between 1975 and 2021 and conducted in several countries worldwide. Among the 56 selected studies, 28 assessed mucosal disorders (rhinitis or rhinosinusitis), 26 evaluated olfactory alterations, and 2 articles addressed both topics [15,16]. Table 1 highlights the main results concerning sinonasal mucosa disorders, while Table 2 reports publications on olfactory dysfunction. The included studies were published between 1975 and 2021 and conducted in several countries worldwide. Among the 56 selected studies, 28 assessed mucosal disorders (rhinitis or rhinosinusitis), 26 evaluated olfactory alterations, and 2 articles addressed both topics [15,16]. Table 1 highlights the main results concerning sinonasal mucosa disorders, while Table 2 reports publications on olfactory dysfunction. -Lower endoscopic and questionnaire scores in the irrigation from pre-RT to 6 months after RT - The between-group differences were most obvious at 2 and 3 months after RT -No differences in Lund-Mackay scores between the two groups from pre-RT to 6 months after RT Maxillary sinuses were the most common involved -Advanced T stage, invasion of the nasal cavity, and nasal irrigation (but not CT or RT dose) were positively associated with the incidence of sinusitis after RT - The incidence of sinusitis peaked at 6-9 months after RT and showed a trend toward stabilization after 1 year Before and 3 months, 1, 3, and 5 years after RT -The rate and severity of sinus abnormalities were highest on the third month after RT -There was no significant increase in the incidence of abnormalities on the fifth year after RT compred to pre-treatment - The anterior ethmoid and maxillary sinuses were the most affected sinuses -No significant increase in the score for sinuses with preexisting abnormality on the third month after RT -Advanced T stage (but not RT dose) was positively associated with the incidence of sinus abnormality on the fifth year after RT      Maxillary sinus (n = 10), oropharynx (n = 10), oral cavity (n = 5), paranasal sinus (n = 5), nasopharynx (n = 6), hypopharynx (n = 2), nasal cavity (n = 1), brain (n = 1), skin (n = 1), unknown primary (n = 1), other (n = 2) (stage NR)
The main chemotherapeutic drugs administered in the studies were platinum compounds, pyrimidine compounds, and taxanes. No study evaluated CT alone.
Nasal endoscopy was performed in 10 studies [15,19,24,26,27,31,33,35,36,43]; two of these used the Lund Endoscopic Staging System, assessing the appearance of nasal endoscopy findings (polyps, oedema, discharge, scarring and crusting), with a total score ranging from 0 to 20 [24,35]. A general clinical examination was included in three articles [14,39,42]. A saccharine test was employed in six studies in order to investigate mucociliary clearance rates [17,19,20,23,29,43]. It consisted of the placement of saccharin on the floor of nasal cavity behind the anterior end of the inferior turbinate. Then, the subjects were asked to swallow every 30 s and to report the first change in taste sensation. The time from the placement of saccharin to the perception of sweetness was noted as the mucociliary clearance time.
The percentage of patients with rhinosinusitis after RT ranged from 7% [42] to 86.1% [30]. The most common isolates in the post-RT CRS group were Staphylococcus aureus, followed by Streptocuccus viridans and Pseudomonas aeruginosa [21,25,40].
In four articles, a higher saccharin perception time was found after RT [19,20,23,29]. Kılıç et al. described a higher saccharin perception time in patients receiving RT dose >60 Gy and in patients receiving CT concurrent to RT [29]. On the contrary, Kamel et al. did not find any correlation between RT dose and mucociliary clearance delay time, endoscopic findings, and Lund-Mackay score [19]. Hu et al. found a decreased saccharin transit time after RT [23].
Maxillary sinuses were the most involved, followed by the anterior ethmoid [14,19,22,28,34]. Park et al. described a higher bilateral CRS percentage in non-RT group (85.7% vs. 60%), and, although RT itself was not associated with sinus surgery, concurrent CT was significantly associated with the need for surgery [14]. Advanced T stage (but not RT dose) was positively associated with the incidence of sinus abnormality in the fifth year after RT [22,28]. In the study by Lu et al., patients in the RT-alone and any-RT groups exhibited an increased risk of CRS compared to patients in the no-RT group (hazard ratio: 6.76 and 2.91, respectively) [42].
The incidence of sinusitis peaked at 3-9 months after RT and showed a trend toward stabilization after 1 year [28,34]. Nasal irrigation reduced CRS post-RT in patients with nasopharyngeal carcinoma [24]. Moreover, higher incidence of CRS after RT was observed in patients who used a nasal sprayer instead of a nasal irrigator [30]. Fewer nasal complaints (overall symptoms, blocked nose, and headache), better quality of life, and less severe endoscopic findings were found in the steroid group at 3 and 6 months after RT [35].
Two articles did not find any significant influence of the RT delivery method for any type of complication [31,33]. Choanal stenosis was found in 4.3% to 23% of cases after RT [23,26,31,38] and negatively affected quality of life [31].
Histologically, areas of ciliary loss, intercellular and intracellular vacuolation, and ciliary dysmorphism, and a decreased number of submucosal gland openings and ciliary areas were found after RT [18,23]. Stoddard et al. found a higher percentage of neutrophilic inflammation and squamous or mucous cell metaplasia in the study group without cytological atypia; furthermore, there was no correlation between cytological changes and symptoms, endoscopic findings, age, smoking, and tumor stage [40].
Nasal cytology showed a radiation-induced rhinitis with neutrophils [15,33]. Furthermore, mucous cell metaplasia appeared in patients during RT [15,33,37]. Mucous cell metaplasia was also found in 20% of laryngectomized patients [36]. Increased squamous metaplasia and subepithelial edema and a higher Lund-Mackay score were observed in radiation-induced CRS compared to CRS without nasal polyps [37]. Riva et al. did not find any correlation between cytological changes and symptoms, endoscopic findings (turbinate hypertrophy, mucosal hyperemia, nasal secretions), age, smoking, tumor stage, and adjuvant RT after total laryngectomy [36]. Yin et al. showed that the patients who received IMRT at a dose less than the threshold had the least damaged nasal mucosa morphology, and functional impairment scores were highest 3 months after RT, with a significant relationship between the turbinate thickness ratio and the radiation dose [43].
The evaluation of CT alone was present in only one paper [59]. The main chemotherapeutic agents administered in the studies were: platinum compounds, pyrimidine compounds, and taxanes. The agent administered for ECT was Bleomycin.
Twenty articles reported substantial smell deterioration, with specific differences in odor identification and odor discrimination. ODT significantly decreased during RT, and its baseline levels had still not recovered at different months after the treatment [44,46,47,52,[55][56][57]63,67], such as OD in two papers [48,63]. In two cases, ODT did not show variations during and after the radiotherapy treatment [48,64]; similarly, OD did not change in the other two papers [47,57]. OI decreased in three papers [52,56,63], while in two papers, it did not show substantial differences before and after treatment [47,57]. A significant reduction of OI was observed in one study during RT, but showed a partial recovery at 3 months follow-up [67]. On the contrary, in another publication, OI was stable at 2-6 weeks after the beginning of radiotherapy, but showed a decrease after treatment at a long term evaluation [48].
Four studies reported an impairment of quality of life [60][61][62]67]. Smell disorders were predictors of depression and anxiety [62], and had a positive correlation with the VAS scale [61].

Discussion
Head and neck cancer treatment is often challenging and multimodal, including radiation therapy and chemotherapy [4]. In recent decades, the development of new radiotherapy techniques has guaranteed a reduction in OARs complications. In particular, RT technology has developed from 3D-conformal planning to IMRT, and the amount of radiation applied to the surrounding tissue has been minimized [7]. However, since the percentage of long-term HNC survivors has been increasing [1], higher attention should be paid to late side effects.
The aim of this review was to summarize the current knowledge of sinonasal side effects of chemotherapy and/or radiation therapy for HNC, with a specific focus on mucosal and smell disorders. Indeed, these complications are often overlooked by physicians, but may negatively affect patients' quality of life.
The prevalence of sinonasal mucosa disorders after RT for HNC is very heterogeneous among studies, ranging from 8% to 86.1%. The 5-year incidence of post-irradiation CRS reached 16.7% in the study by Hsin et al. conducted on 102 patients with nasopharyngeal carcinoma (NPC) [32]. However, the applicability of this information is limited because this study investigated only patients with NPC, and not with the HNC of other sites. Furthermore, potential confounders such as CT or previous sinonasal disorders have not been properly considered or controlled. Radiation-induced choanal stenosis is a rare late toxicity, observed in 4.3% to 23% of cases after RT [23,26,31,38]. It leads to serious difficulty in breathing through, and discharge from, the nose. When it comes to improving these symptoms, transnasal endoscopic surgery is often needed [31].
The results obtained by Lu et al. revealed that the risk of CRS was significantly higher in the RT groups compared to the no-RT group. In particular, the 5-year cumulative incidence rates of CRS in the RT-alone, any-RT, and no-RT groups were estimated as 12%, 9.3%, and 4.5%, respectively. The hazard ratios for CRS were 6.76 and 2.91 in the RT-alone and any-RT groups, respectively, compared to the no-RT group. Moreover, the HNC site is another factor associated with the risk of post-treatment CRS: the patients with NPC had the highest incidence of CRS (HR 4.54), while patients with oral cancer had the lowest (HR 0.29) [42].
The incidence of rhinosinusitis peaked at 3-9 months after RT and showed a trend toward stabilization after one year [27,34]. The highest incidence and severity of sinus mucosa disease were found at 3 months after RT (about 67% of cases), which decreased gradually over time. Advanced tumor stage and smoking habit were predisposing factors for sinus mucosa disease, while age, sex, RT dose, and nodal status were not. On the contrary, no factors could predict sinus mucosa disease improvement after RT [22]. In the study by Su et al., a lot of patients with CRS before treatment suffered aggravated symptoms after RT, but 75.3% of patients without CRS before RT developed it after RT. Advanced tumor stage, invasion of the nasal cavity, and nasal irrigation, but not CT or RT dose, were positively associated with the incidence of rhinosinusitis after RT [28]. In the paper by Hamilton et al., nasal crusting, CRS, and epistaxis were quite common (nasal crusting in 16% of patients, epistaxis in 16%, and chronic sinusitis in 8% of cases), attributable to the higher proportion of subjects treated with high dose RT to the nasopharynx in their study [39].
The histopathological features of radiation-induced CRS are different from non-post-RT CRS. Kuhar et al. observed that patients with RT-induced CRS exhibited greater squamous metaplasia and subepithelial edema compared to patients with CRSsNP, and decreased eosinophilia and basement membrane thickening compared to patients with CRSwNP [37]. Riva et al. showed that a radiation-induced rhinitis with neutrophils and sometimes bacteria occurred in 70% of cases and persisted after 1 month; mucous cell metaplasia appeared in 10% of patients during RT and disappeared after 3 months, and squamous cell metaplasia was observed in 10% of cases, only after the end of RT [15]. These results are in agreement with their previous study on the late effects of RT for nasopharyngeal cancer that showed 40% of neutrophilic rhinitis, 20% squamous cell metaplasia, and 13% mucous cell metaplasia, after a median follow-up of 59 months [33]. The most common isolates in the post-RT CRS group were Staphylococcus aureus, followed by Streptocuccus viridans and Pseudomonas aeruginosa [21,25,40].
Few studies have investigated the relationship between RT dose and radiation-induced damage to the nasal mucosa. Indeed, RT dose to nasal cavities was detected only in three studies, and ranged from 13.59 to 56.38 Gy [15,18,43]. Therefore, there is insufficient information on the radiation tolerance of the nasal mucosa. Yin et al. investigated 66 patients and assessed the radiation tolerance of the nasal mucosa by performing the modified saccharin test, endoscopy test, MRI, and a SNOT-20 survey. The results showed that there was a threshold radiation dose, and that above such an RT dose, nasal tissues may not recover from radiation-induced damage. The threshold doses of IMRT ranged from 37 to 40 Gy. A low dose of IMRT (inferior to the threshold dose) was associated with higher mucocilia transport rate, better endoscopy test score, and improved SNOT-20 score [43]. Moreover, an association between turbinate thickness ratio and radiation dose was observed. Riva et al. noticed a significant correlation between mean dose (Dmean) and near maximum dose (D2%) to inferior turbinates and neutrophilic rhinitis, and between D2% to inferior turbinates and mucous cell metaplasia at the end of RT [15].
Treatment for radiation-induced CRS was reported by four studies, and included functional endoscopic sinus surgery [14,23,27,37]. An algorithm for radiation-induced CRS therapy, including medical and surgical options, has never been proposed.
Smell perception plays an important role in life experiences and influences every aspect. Olfactory dysfunction can limit daily life activities and have an adverse effect on nutritional status. Concerning HNC, olfactory disorders may be secondary to direct damage caused by tumors, or to treatments, including surgery, RT, and CT. In particular, RT and CT can alter smell perception by damaging the olfactory epithelium and/or nerves.
The percentage of patients with olfactory impairment after RT ranged from 7% to 76%. Smell disorders appeared during RT and decreased after treatment, remaining higher than non-irradiated patients [45,51,56,57,[61][62][63]65,66,68]. Studies that did not find any differences in smell perception between irradiated and non-irradiated patients mainly included subjects affected by non-nasopharyngeal tumors [50,53,69]. Therefore, this lack of difference could be explained by the fact that olfactory epithelium was outside the radiation fields in these patients.
A worse olfactory function was observed in patients receiving 'high dose' compared to 'low dose' to the olfactory epithelium. After therapy, 40% and 7% reported subjective olfactory decline in high and low RT dose groups, respectively [52].
A small part of cited studies was derived from 2019-2022, which coincides with the SARS-CoV-2 pandemic. Since dysfunction of smell and taste could be an effect of SARS-CoV-2 infection, it may partially modulate results in the most recent studies.
Only two studies described both sinonasal mucosal disorders and olfactory disfunction after RT treatment [15,16]. Wang et al. performed patients' evaluations before and 12 months after RT, observing an increase in sinonasal mucosa and smell disorders after treatment. Moreover, olfactory alterations correlated with total and ethmoid Lund-Mckay scores [16]. Riva et al. evaluated the patients before, during, and after RT (3 months). In agreement with Wang et al., a concurrent increase in mucosal and olfactory disorders was observed during RT. Furthermore, Riva et al. found that nasal symptoms and endoscopic findings peaked at the end of RT [15]. Further studies are necessary to identify where OAR should be set in order to reduce the incidence of sinonasal side effects.
The impact of CT on nasal side effects has been poorly analyzed. Only one study evaluated olfactory complaints during CT for HNC. The TDI score decreased during the second CT cycle, especially in older patients (>55 years), and reached almost its initial levels after 3 weeks of recovery time [59]. On the other hand, the impact of CT alone on sinonasal mucosa has never been investigated.

Conclusions
Sinonasal mucosa and smell disorders are a common post-treatment side effect of CT and/or RT in HNC patients. The incidence and severity of olfactory dysfunction and chronic rhinosinusitis were highest at the end of RT and at 3 months after treatment and decreased gradually with time. Smell acuity deterioration and chronic rhinosinusitis after RT seemed related to radiation dose on olfactory area and nasal cavities, but different degrees of recovery were observed. Therefore, it is important to establish the severity of chronic rhinosinusitis and olfactory dysfunction in order to find strategies to support patients and improve their quality of life. Further studies are necessary to better assess the role of medical and surgical treatments of sinonasal side effects of CT and/or RT for HNC. Finally, the role of CT should not be overlooked, and future studies are mandatory to assess its effect on nasal cavities and paranasal sinuses.

Conflicts of Interest:
The authors declare no conflict of interest.